The objective of this competitive renewal is to define the role of TNF in murine anti-tumor responses. We will also define the cellular basis of actual and potential TNF-producers at the population and organ level under basal conditions or after stimulation in vivo and in vitro. The in vitro tumor lysis and in vivo-induced local tumor necrosis are two related and well defined activities of TNF. However, most of the animal and human studies showed that administration of recombinant TNF to tumor-bearing hosts had very limited therapeutic effects. In this grant we plan to use three new approaches to define the role of TNF in anti-tumor responses. 1) We will determine if treatment with antibodies that neutralize TNF lytic functions will affect tumor development and behavior. Thus, answering the question on whether TNF has anything to do with anti-tumor responses in vivo. 2) We developed models in which TNF production is selectively induced in vivo. We will determine if these increased levels of soluble TNF and of TNF-producing cells will affect tumor development and behavior. This approach may answer the question on whether endogenous production of TNF rather than exogenous administration offers a therapeutic advantage in anti-tumor responses. And 3) We will determine if membrane-associated TNF (Ma-TNF) rather than soluble TNF (So-TNF) may have a preferential role in the anti-tumor effects. A preferential role of Ma-TNF rather than So-TNF, may also explain the therapeutic failures with exogenous So-TNF. The main in vivo models to be tested are tumor development after chemical carcinogens and behavior of transplanted tumors in syngeneic mice. Mice in both types of models will be treated using protocols that inhibit TNF activity or that induce TNF production in vivo. These in vivo models will be further developed and refined using in vitro triggering of TNF production by T and non-T lymphoid cells by various agents. Our studies will also include the analysis of the in vivo and in vitro behavior of tumor variants selected for high sensitivity or for resistance to TNF in vitro killing. It is felt that the proposed approaches will clarify our understanding of the biology of TNF and its possible use in anti-tumor therapy.